Pith. sign in

REVIEW 3 major objections 5 minor 73 references

Support matrix machine: exploring sample sparsity, low rank, and adaptive sieving in high-performance computing

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims an augmented Lagrangian with semismooth Newton-CG solves large-scale support matrix machines by exploiting sample sparsity and low-rank solutions, reducing per-iteration cost from $O(npq)$ to $O(pq\max\{|J_1|, |\alpha|\})$.

desk verdict A useful algorithmic extension with impressive experiments, but the advertised per-iteration complexity omits the O(npq) residual formation, so the main theoretical claim needs correction. read the letter →

arxiv 2412.08023 v1 pith:BZQIFESO submitted 2024-12-11 math.OC

Can Wu , Dong-Hui Li , Defeng Sun This is my paper · ORCID
classification math.OC MSC 90C0690C2590C90
keywords supportmatrixmachinesamplesparsitylow-rankregularizationadaptivesievingaugmentedLagrangianmethodsemismoothNewtonnuclearnormlarge-scaleclassification
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that the support matrix machine---a classifier for matrix-shaped data with Frobenius and nuclear-norm regularization plus hinge loss---can be solved when $n$ is hundreds of thousands to a million, not just small-sample problems. The proposed method is an augmented Lagrangian outer loop whose subproblems are solved by a semismooth Newton-CG inner solver, and the central claim is that the inner linear algebra costs $O(pq\max\{|J_1|, |\alpha|\})$ per iteration instead of $O(npq)$, where $J_1$ is the active-support-matrix set and $|\alpha|$ is the rank of the solution matrix $W$. The paper also develops an adaptive sieving strategy that reuses active samples from one grid value of $C$ to warm-start the next, with finite termination. If the claim is correct, cross-validation over $C$ and other model-selection tasks become feasible on large matrix datasets.

What carries the argument

The machinery is the pair of index sets $J_1$ and $\alpha$, together with the generalized Jacobians of the two proximal or projection maps. $J_1$ collects training samples whose dual slack is strictly inside $(0,C)$---the active support matrices---and $\alpha$ collects singular values of the projection input that exceed $\tau$, so $|\alpha|$ equals the rank of the optimal $W$. Choosing $M\in\partial\Pi_S$ supported on $J_1$ and $G\in\partial\Pi_{\mathcal{B}_2^\tau}$ with the sparse structure induced by $\alpha$ makes the Newton operator $\tilde V$ act only on $|J_1|$ samples and an $|\alpha|$-dimensional spectral block; the augmented Lagrangian outer loop and the adaptive sieving strategy then reduce the number and size of the subproblems solved.

What would settle it

Choose a dataset in which a large fraction of training matrices lie exactly on the optimal margin and take $\tau$ small enough that many singular values of $-A^*\lambda$ exceed $\tau$; then measure the time per CG iteration as $n$ grows. If the cost scales like $npq$ rather than $\max\{|J_1|,|\alpha|\}pq$, the advertised complexity reduction is not holding in that regime.

Watch

Extended reading notes

Core claim

At a KKT point of the SMM model, the dual variable satisfies $\lambda\in[0,C]^n$ and the regression matrix has the form $W=-\sum_{j=1}^n \lambda_j y_j X_j-\Lambda$, so only samples with positive dual weight contribute to $W$; samples with $0<(-\lambda)_j<C$ are the active support matrices and lie on the margin. In parallel, the optimality relation $W=\mathrm{Prox}_{\tau\|\cdot\|_*}(-A^*\lambda)$ shows that the rank of $W$ equals the number of singular values of $-A^*\lambda$ exceeding $\tau$, denoted $|\alpha|$. The paper's discovery is that these two sets are exactly the sparsity needed by the semismooth Newton-CG solver: the generalized-Jacobian factor $M$ is supported on $J_1$, and the spectral-ball factor $G$ has a block-sparse structure indexed by $\alpha$, so the dominant matrix-vector products in the Newton system cost $O(|J_1|pq)$ and $O(|\alpha|pq)$ instead of $O(npq)$. Under a strict-complementarity condition the ALM iterates converge R-superlinearly, and the SNCG inner loop converges superlinearly when $J_1$ is nonempty.

Load-bearing premise

The speed and superlinear convergence depend on the active support set and the solution rank being genuinely small; if a dataset produces many active support matrices or a high-rank solution, the per-iteration cost climbs back toward $O(npq)$ and the reported gains shrink.

Editorial extensions

If this is right

  • ALM-SNCG solves SMM instances with $n$ up to $10^6$ and $pq$ in the tens of thousands, where F-ADMM runs out of memory or cannot reach the requested accuracy within two hours.
  • On the four real datasets tested, ALM-SNCG is on average 422.7 times faster than F-ADMM at medium accuracy and 477.2 times faster at high accuracy.
  • Adaptive sieving generates a $C$-grid solution path with average speedups of 2.69 and 3.07 over warm-started ALM on synthetic data, and 1.50 and 1.92 on MNIST.
  • The same algorithm framework carries over to convex SMM variants such as multiclass, weighted, transfer, and pinball SMMs, because the active-sample and low-rank structure is inherited from the same KKT system.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: tracking $|J_1|/n$ and $|\alpha|/p$ online would let a user detect when the $O(pq\max\{|J_1|,|\alpha|\})$ promise is about to fail, since the paper's own cost analysis makes the dependence explicit.
  • Editorial inference: the adaptive sieving idea should transfer to the convex SMM variants listed in the paper, but the transfer is an extrapolation until experiments are run.
  • Editorial inference: combining the sample sieve with a feature or subspace sieve is the natural next step, and a concrete test is whether the two reductions compound on high-resolution image datasets where $pq$ is large.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper develops an augmented Lagrangian method with a semismooth Newton-CG inner solver (ALM-SNCG) for the support matrix machine (model (1)). It exploits two structural properties at KKT points: sample sparsity, expressed through the active support-matrix index set J1, and low rank of the regression matrix W, expressed through the index set α. These are used to reduce the cost of the Hessian-vector product in the Newton linear system from O(npq) to O(pq max{|J1|, |α|}). The paper also proposes an adaptive sieving (AS) strategy for generating solution paths over a grid of C values and proves its finite termination. Numerical experiments on synthetic data with up to one million samples and on four real datasets report substantial speedups over F-ADMM and isPADMM, including an average speedup of 422.7x over F-ADMM on real data.

Significance. If the advertised complexity and convergence results are taken at face value, the paper is a useful step toward making SMM tractable at large sample sizes. The extension of adaptive sieving to sample sparsity is nontrivial, and the experimental scope (up to one million samples) is impressive. The paper also provides concrete KKT-residual stopping criteria and convergence-rate statements, and it includes proofs for the quadratic-growth condition and the finite termination of the AS strategy. However, the advertised per-iteration complexity covers only one Hessian-vector product, not the complete Newton iteration, and the numerical comparison with F-ADMM and isPADMM is weakened by an unexplained warm-start procedure for ALM-SNCG. The contribution is therefore conditionally significant: the algorithmic ideas are sound, but the end-to-end complexity and the fairness of the speedup comparisons need to be established.

major comments (3)
  1. [Section 4, Table 1; Eq. (17)] The complexity analysis counts only the cost of applying the operator eV in (32) to a direction H inside the CG loop. It does not count the cost of forming the residual grad phi_k in (17) at each Newton step and at each line-search trial. Forming grad phi_k requires the vector omega = -lambda_k - sigma_k(AW + b y - e_n) for all n samples (definition (11)) and the adjoint application A^* Pi_S(omega). For a new W, AW costs O(npq), and Pi_S(omega) is supported on all support matrices, not only on J1; Figure 1 reports support-matrix fractions up to 86%. No incremental update of AW is described. Therefore the paper does not establish an O(pq max{|J1|, |alpha|}) cost per iteration of Algorithm 2; it establishes that cost only for a single CG matvec. The speedup claims in Section 6.2.2 need a full per-iteration complexity analysis or a profiling/timing breakdown that separates residual formation, line search, and CG matvec costs.
  2. [Section 6.2] The description of the experimental setup states that ALM-SNCG is initialized with a lower-accuracy starting point obtained from isPADMM, using up to four iterations on synthetic data and up to ten on real data, while the other methods start from the origin. The wall-clock time spent in those initial isPADMM iterations is not reported, and the paper does not state whether it is included in the ALM-SNCG times. If it is not included, the speedup ratios in Tables 4 and 5 are not a like-for-like comparison. Please report the warm-start overhead explicitly, run all methods from the same starting point, or present cumulative times that include the cost of generating the warm start.
  3. [Section 3.2, Theorem 2, Proposition 2] The superlinear convergence of the inner SNCG method is conditional on the nonemptiness of the index set J1 at the limit, which is equivalent to constraint nondegeneracy. The numerical sections report values of |J1| but do not state whether this condition holds in the tested configurations. Since the advertised high efficiency is partly attributed to the fast local convergence of SNCG, the manuscript should either verify the condition for the reported instances or provide a fallback convergence statement for cases where J1 is empty or very small.
minor comments (5)
  1. [Section 1, contribution 2] The sentence claiming to reduce the cost 'from O(pq max{|J1|, |alpha|}) to O(npq)' reverses the direction; it should read 'from O(npq) to O(pq max{|J1|, |alpha|})'.
  2. [Table 2 and Tables 4-5] There are typos: 'Tabel' in Table 2, and 'exibited' and 'achive' in the text around Tables 4 and 5.
  3. [Proposition 1, Eq. (14)] The symbol I is used both for the identity matrix and for the matrix I := U [I_p 0] V^T; please use a different symbol, such as \bar{I}, to avoid confusion.
  4. [Theorem 1] The constant beta is defined with an unspecified parameter gamma >= 1; please state the role of gamma or give a precise reference for this constant.
  5. [Figure 1] The caption uses NSM without defining it; please define support matrices, active support matrices, and non-support matrices in the caption or in the surrounding text.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the ALM-SNCG convergence, the per-CG-iteration cost reduction, and the adaptive sieving strategy are derived from KKT conditions and established theorems, not from the conclusions being proved.

full rationale

I walked the derivation chain from the SMM model (1) through the dual problem (D), the KKT system (3), the ALM subproblem (8), the semismooth Newton-CG method in Algorithm 2, and the adaptive sieving Algorithm 3. No load-bearing step reduces to its own input by definition. The sample-sparsity and low-rank properties are characterized independently: sample sparsity follows from W = -A*lambda - Lambda and the dual feasibility -lambda in [0,C]^n, while low rank follows from W = Prox_{tau||.||_*}(-A*lambda) and the singular-value thresholding formula. The J1 and alpha sets are defined at each iterate and are only identified with active support matrices and the solution rank after the KKT relations (30); they are not fitted quantities used as predictions. The adaptive sieving strategy is not circular: even though it initializes from the previous grid point's support set, Algorithm 3 explicitly scans the unselected samples, computes their margins, and adds any sample with v_j >= 0. Its finite-termination proof in Appendix C shows that termination with J^k(C_i) = empty supplies exactly the KKT conditions of the full problem; this is a stopping certificate, not a hidden assumption of the desired output. The self-citations to [9], [38], [63], [66], [67], and [69] are to established convergence theorems and prior adaptive-sieving frameworks, invoked as lemmas with independent proofs rather than as the source of the SMM-specific claims. The complexity statement in Section 4 and Table 1 is explicitly scoped to the cost of applying the operator eV inside the CG step; whether the full residual formation and line-search costs should be counted in the advertised per-iteration complexity is a legitimate completeness or correctness question, but it is not a circularity. The empirical speedups are benchmark comparisons against F-ADMM and isPADMM and do not depend on the paper's own outputs as ground truth. Overall, the paper's derivation is self-contained for its algorithmic claims, with only minor, non-load-bearing reliance on the authors' prior work.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central algorithm is a direct application of known ALM and SNCG theory; the only new postulates are problem-specific structural assumptions (strict complementarity and small active support set). No new physical entities are introduced.

free parameters (2)
  • dmax = 500
    Maximum number of newly added samples per adaptive sieving round in Algorithm 3; hand-chosen, affects speed not convergence.
  • bε = 0.05 / 0.1 / 0.4
    Safety margin that expands initial index sets in Algorithm 3, tuned per dataset (Section 6.1); affects speed and solution path quality.
assumptions (5)
  • domain assumption The dual quadratic growth condition (12) holds at the solution, ensured by the strict complementarity condition rank(W)+rank(τI−Λ)=p.
    Required by Theorem 1 for R-superlinear convergence of ALM; not guaranteed for all SMM instances.
  • domain assumption Constraint nondegeneracy at the ALM subproblem solution, equivalent to nonempty active support matrix set J1(-λ̂).
    Required by Proposition 2 and Theorem 2 for superlinear convergence of SNCG; if J1 is empty, only weaker convergence is assured.
  • standard math The KKT system (3) has a nonempty solution set because the SMM objective is lower level-bounded and constraints are linear.
    Invoked in Section 2 via standard convex analysis results.
  • domain assumption The training set contains at least one positive and one negative label.
    Stated in Section 2 to avoid degenerate classification.
  • standard math Projection operators onto S and the spectral ball are strongly semismooth and their Clarke Jacobians can be characterized as in Appendix B.
    Used to construct the generalized Jacobian in Section 3.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Support matrix machine: exploring sample sparsity, low rank, and adaptive sieving in high-performance computing." pith.science (2026). https://pith.science/paper/BZQIFESO

@misc{pith2026241208023,
  author       = {Pith},
  title        = {Pith review of: Support matrix machine: exploring sample sparsity, low rank, and adaptive sieving in high-performance computing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BZQIFESO}},
  note         = {Machine review of arXiv:2412.08023}
}
read the original abstract

Support matrix machine (SMM) is a successful supervised classification model for matrix-type samples. Unlike support vector machines, it employs low-rank regularization on the regression matrix to effectively capture the intrinsic structure embedded in each input matrix. When solving a large-scale SMM, a major challenge arises from the potential increase in sample size, leading to substantial computational and storage burdens. To address these issues, we design a semismooth Newton-CG (SNCG) based augmented Lagrangian method (ALM) for solving the SMM. The ALM exhibits an asymptotic R-superlinear convergence if a strict complementarity condition is satisfied. The SNCG method is employed to solve the ALM subproblems, achieving at least a superlinear convergence rate under the nonemptiness of an index set. Furthermore, the sparsity of samples and the low-rank nature of solutions enable us to reduce the computational cost and storage demands for the Newton linear systems. Additionally, we develop an adaptive sieving strategy that generates a solution path for the SMM by exploiting sample sparsity. The finite convergence of this strategy is also demonstrated. Numerical experiments on both large-scale real and synthetic datasets validate the effectiveness of the proposed methods.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

73 extracted references · 69 canonical work pages

  1. [1]

    Artacho, F.A., Geoffroy, M.H.: Characterization of metric regularity of subdifferentials. J. Convex Anal. 15(2), 365–380 (2008)

    work page 2008

  2. [2]

    Bauschke, H.H., Borwein, J.M., Li, W.: Strong conical hull intersection property, bounded linear regularity, Jameson’s property (G), and error bounds in convex op- timization. Math. Program. 86(1), 135–160

    work page

  3. [3]

    Chen, L., Sun, D.F., Toh, K.C.: An efficient inexact symmetric Gauss-Seidel based majorized ADMM for high-dimensional convex composite conic programming. Math. Program. 161(1), 237–270 (2017)

    work page 2017

  4. [4]

    Chen, Y., Hang, W., Liang, S., Liu, X., Li, G., Wang, Q., Qin, J., Choi, K.S.: A novel transfer support matrix machine for motor imagery-based brain computer interface. Front. Neurosci. 14, 606949 (2020)

    work page 2020

  5. [5]

    John Wiley and Sons, New York (1983)

    Clarke, F.H.: Optimization and Nonsmooth Analysis. John Wiley and Sons, New York (1983)

    work page 1983

  6. [6]

    Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20, 273–297 (1995)

    work page 1995

  7. [7]

    Cui, Y., Ding, C., Zhao, X.: Quadratic growth conditions for convex matrix optimization problems associated with spectral functions. SIAM J. Optim. 27(4), 2332–2355 (2017)

    work page 2017

  8. [8]

    arXiv preprint arXiv:1610.00875 (2016)

    Cui, Y., Sun, D.F., Toh, K.C.: On the asymptotic superlinear convergence of the aug- mented Lagrangian method for semidefinite programming with multiple solutions. arXiv preprint arXiv:1610.00875 (2016)

    arXiv 2016

Show all 73 references
  1. [9]

    Cui, Y., Sun, D.F., Toh, K.C.: On the R-superlinear convergence of the KKT residuals generated by the augmented Lagrangian method for convex composite conic program- ming. Math. Program. 178(1), 381–415 (2019)

    2019

  2. [10]

    Duan, B., Yuan, J., Liu, Y., Li, D.: Quantum algorithm for support matrix machines. Phys. Rev. A 96(3), 032301 (2017)

    2017

  3. [11]

    Springer Science & Business Media, New York (2007)

    Facchinei, F., Pang, J.S.: Finite-Dimensional Variational Inequalities and Complemen- tarity Problems. Springer Science & Business Media, New York (2007)

    2007

  4. [12]

    Neural Comput

    Feng, R., Xu, Y.: Support matrix machine with pinball loss for classification. Neural Comput. Appl. 34, 18643–18661 (2022)

    2022

  5. [13]

    Feng, R., Zhong, P., Xu, Y.: A subspace elimination strategy for accelerating support matrix machine. Pac. J. Optim. 18(1), 155–176 (2022)

    2022

  6. [14]

    Geng, M., Xu, Z., Mei, M.: Fault diagnosis method for railway turnout with pinball loss-based multiclass support matrix machine. Appl. Sci. 13(22), 12375 (2023)

    2023

  7. [15]

    Goldstein, T., O’Donoghue, B., Setzer, S., Baraniuk, R.: Fast alternating direction op- timization methods. SIAM J. Imaging Sci. 7(3), 1588–1623 (2014)

    2014

  8. [16]

    Com- put

    Hang, W., Feng, W., Liang, S., Wang, Q., Liu, X., Choi, K.S.: Deep stacked support matrix machine based representation learning for motor imagery eeg classification. Com- put. Meth. Programs Biomed. 193, 105466 (2020)

    2020

  9. [17]

    Hang, W., Li, Z., Yin, M., Liang, S., Shen, H., Wang, Q., Qin, J., Choi, K.S.: Deep stacked least square support matrix machine with adaptive multi-layer transfer for EEG classification. Biomed. Signal Process. Control 82, 104579 (2023)

    2023

  10. [18]

    Harrow, A.W., Hassidim, A., Lloyd, S.: Quantum algorithm for linear systems of equa- tions. Phys. Rev. Lett. 103(15), 150502 (2009)

    2009

  11. [19]

    Hastie, T., Tibshirani, R., Friedman, J.H., Friedman, J.H.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction, vol. 2. Springer (2009) Support matrix machine: exploring sample sparsity... 37

    2009

  12. [20]

    In: International Conference on Machine Learning, pp

    Hsieh, C.J., Olsen, P.: Nuclear norm minimization via active subspace selection. In: International Conference on Machine Learning, pp. 575–583. PMLR (2014)

    2014

  13. [21]

    Jiang, K.: Algorithms for Large Scale Nuclear Norm Minimization and Convex Quadratic Semidefinite Programming Problems. Ph.D. thesis (2011)

    2011

  14. [22]

    In: Discrete Geometry and Optimization, pp

    Jiang, K., Sun, D.F., Toh, K.C.: Solving nuclear norm regularized and semidefinite matrix least squares problems with linear equality constraints. In: Discrete Geometry and Optimization, pp. 133–162. Springer (2013)

    2013

  15. [23]

    Neural Networks (2024)

    Kumari, A., Akhtar, M., Shah, R., Tanveer, M.: Support matrix machine: A review. Neural Networks (2024). https://doi.org/10.1016/j.neunet.2024.106767

    2024

  16. [24]

    Laurent, E.G., Vivian, V., Tarek, R.: Safe feature elimination in sparse supervised learn- ing. Pac. J. Optim. 8(4), 667–698 (2012)

    2012

  17. [25]

    Li, H., Xu, Y.: Support matrix machine with truncated pinball loss for classification. Appl. Soft. Comput. 154, 111311 (2024)

    2024

  18. [26]

    Li, Q., Jiang, B., Sun, D.F.: MARS: A second-order reduction algorithm for high- dimensional sparse precision matrices estimation. J. Mach. Learn. Res. 24, 1–44 (2023)

    2023

  19. [27]

    IEEE Trans

    Li, X., Cheng, J., Shao, H., Liu, K., Cai, B.: A fusion cwsmm-based framework for rotating machinery fault diagnosis under strong interference and imbalanced case. IEEE Trans. Industr. Inform. 18(8), 5180–5189 (2022)

    2022

  20. [28]

    IEEE Trans

    Li, X., Cheng, J., Shao, H., Liu, K., Cai, B.: A fusion CWSMM-based framework for rotating machinery fault diagnosis under strong interference and imbalanced case. IEEE Trans. Ind. Inform. 18(8), 5180–5189 (2022)

    2022

  21. [29]

    Li, X., Li, S., Wei, D., Si, L., Yu, K., Yan, K.: Dynamics simulation-driven fault diagnosis of rolling bearings using security transfer support matrix machine. Reliab. Eng. Syst. Saf. 243, 109882 (2024)

    2024

  22. [30]

    Li, X., Li, Y., Yan, K., Shao, H., Lin, J.J.: Intelligent fault diagnosis of bevel gearboxes using semi-supervised probability support matrix machine and infrared imaging. Reliab. Eng. Syst. Saf. 230, 108921 (2023)

    2023

  23. [31]

    IEEE ASME Trans

    Li, X., Li, Y., Yan, K., Si, L., Shao, H.: An intelligent fault detection method of indus- trial gearboxes with robustness one-class support matrix machine toward multisource nonideal data. IEEE ASME Trans. Mechatron. 29(1), 388–399 (2024)

    2024

  24. [32]

    IEEE Trans

    Li, X., Shao, H., Lu, S., Xiang, J., Cai, B.: Highly efficient fault diagnosis of rotating machinery under time-varying speeds using LSISMM and small infrared thermal images. IEEE Trans. Syst. Man Cybern. Syst. 52(12), 7328–7340 (2022)

    2022

  25. [33]

    Li, X., Sun, D.F., Toh, K.C.: QSDPNAL: A two-phase augmented Lagrangian method for convex quadratic semidefinite programming. Math. Program. Comput. 10, 703–743 (2018)

    2018

  26. [34]

    Li, X., Yang, Y., Pan, H., Cheng, J., Cheng, J.: Non-parallel least squares support matrix machine for rolling bearing fault diagnosis. Mech. Mach. Theory 145, 103676 (2020)

    2020

  27. [35]

    IEEE Trans

    Li, Y., Wang, D., Liu, F.: The auto-correlation function aided sparse support matrix machine for EEG-based fatigue detection. IEEE Trans. Circuits Syst. II-Express Briefs 70(2), 836–840 (2023)

    2023

  28. [36]

    IEEE Trans

    Liang, S., Hang, W., Lei, B., Wang, J., Qin, J., Choi, K.S., Zhang, Y.: Adaptive multi- model knowledge transfer matrix machine for EEG classification. IEEE Trans. Neural Netw. Learn. Syst. 35(6), 7726–7739 (2024)

    2024

  29. [37]

    Liang, S., Hang, W., Yin, M., Shen, H., Wang, Q., Qin, J., Choi, K.S., Zhang, Y.: Deep EEG feature learning via stacking common spatial pattern and support matrix machine. Biomed. Signal Process. Control 74, 103531 (2022)

    2022

  30. [38]

    arXiv preprint arXiv:2009.08719 (2020)

    Lin, M., Yuan, Y., Sun, D.F., Toh, K.C.: Adaptive sieving with PPDNA: Generating solution paths of exclusive lasso models. arXiv preprint arXiv:2009.08719 (2020)

    2020 arXiv

  31. [39]

    Liu, Y.J., Sun, D., Toh, K.C.: An implementable proximal point algorithmic framework for nuclear norm minimization. Math. Program. 133, 399–436 (2012)

    2012

  32. [40]

    In: International Conference on Machine Learning, pp

    Luo, L., Xie, Y., Zhang, Z., Li, W.J.: Support matrix machines. In: International Conference on Machine Learning, pp. 938–947. PMLR (2015)

    2015

  33. [41]

    Mangasarian, O.L.: A simple characterization of solution sets of convex programs. Oper. Res. Lett. 7, 21–26 (1988)

    1988

  34. [42]

    Meng, F., Sun, D.F., Zhao, G.: Semismoothness of solutions to generalized equations and the Moreau-Yosida regularization. Math. Program. 104, 561–581 (2005) 38 C. Wu, D. Li, D. Sun

    2005

  35. [43]

    In: International Conference on Machine Learning, pp

    Ogawa, K., Suzuki, Y., Takeuchi, I.: Safe screening of non-support vectors in pathwise SVM computation. In: International Conference on Machine Learning, pp. 1382–1390. PMLR (2013)

    2013

  36. [44]

    Pan, H., Sheng, L., Xu, H., Tong, J., Zheng, J., Liu, Q.: Pinball transfer support matrix machine for roller bearing fault diagnosis under limited annotation data. Appl. Soft. Comput. 125, 109209 (2022)

    2022

  37. [45]

    Pan, H., Sheng, L., Xu, H., Zheng, J., Tong, J., Niu, L.: Deep stacked pinball transfer matrix machine with its application in roller bearing fault diagnosis. Eng. Appl. Artif. Intell. 121, 105991 (2023)

    2023

  38. [46]

    IEEE Trans

    Pan, H., Xu, H., Zheng, J., Shao, H., Tong, J.: A semi-supervised matrixized graph embedding machine for roller bearing fault diagnosis under few-labeled samples. IEEE Trans. Industr. Inform. 20(1), 854–863 (2024)

    2024

  39. [47]

    Pan, H., Xu, H., Zheng, J., Su, J., Tong, J.: Multi-class fuzzy support matrix machine for classification in roller bearing fault diagnosis. Adv. Eng. Inform. 51, 101445 (2022)

    2022

  40. [48]

    Pan, H., Yang, Y., Zheng, J., Li, X., Cheng, J.: A fault diagnosis approach for roller bearing based on symplectic geometry matrix machine. Mech. Mach. Theory140, 31–43 (2019)

    2019

  41. [49]

    IEEE Trans

    Pan, X., Xu, Y.: A novel and safe two-stage screening method for support vector ma- chine. IEEE Trans. Neural Netw. Learn. Syst. 30(8), 2263–2274 (2019)

    2019

  42. [50]

    Qi, H., Sun, D.F.: A quadratically convergent Newton method for computing the nearest correlation matrix. SIAM J. Matrix Anal. Appl. 28(2), 360–385 (2006)

    2006

  43. [51]

    Qi, L., Sun, J.: A nonsmooth version of Newton’s method. Math. Program. 58, 353–367 (1993)

    1993

  44. [52]

    Qian, C., Tran-Dinh, Q., Fu, S., Zou, C., Liu, Y.: Robust multicategory support matrix machines. Math. Program. 176, 429–463 (2019)

    2019

  45. [53]

    IEEE Trans

    Razzak, I., Blumenstein, M., Xu, G.: Multiclass support matrix machines by maximizing the inter-class margin for single trial EEG classification. IEEE Trans. Neural Syst. Rehabilitation Eng. 27(6), 1117–1127 (2019)

    2019

  46. [54]

    in Mathemat- ical Programming at Oberwolfach, Math

    Robinson, S.M.: Some continuity properties of polyhedral multifunctions. in Mathemat- ical Programming at Oberwolfach, Math. Program. Stud. pp. 206–214 (1981)

    1981

  47. [55]

    Princeton University Press, Princeton (1970)

    Rockafellar, R.T.: Convex Analysis. Princeton University Press, Princeton (1970)

    1970

  48. [57]

    Rockafellar, R.T.: Augmented Lagrangians and applications of the proximal point algo- rithm in convex programming. Math. Oper. Res. 1(2), 97–116 (1976)

    1976

  49. [58]

    Rockafellar, R.T., Wets, R.J.B.: Variational Analysis, vol. 317. Springer Science & Business Media (2009)

    2009

  50. [59]

    John Wiley & Sons (2013)

    Sanei, S., Chambers, J.A.: EEG Signal Processing. John Wiley & Sons (2013)

    2013

  51. [60]

    Shapiro, A.: Sensitivity analysis of generalized equations. J. Math. Sci. 115(4), 2554– 2565 (2003)

    2003

  52. [61]

    In: International Conference on Machine Learning, pp

    Wang, J., Wonka, P., Ye, J.: Scaling SVM and least absolute deviations via exact data reduction. In: International Conference on Machine Learning, pp. 523–531. PMLR (2014)

    2014

  53. [62]

    Linear Alg

    Watson, G.A.: Characterization of the subdifferential of some matrix norms. Linear Alg. Appl. 170, 33–45 (1992)

    1992

  54. [63]

    INFORMS J

    Wu, C., Cui, Y., Li, D., Sun, D.F.: Convex and nonconvex risk-based linear regression at scale. INFORMS J. Comput. 35(4), 797–816 (2023)

    2023

  55. [64]

    Xu, H., Pan, H., Zheng, J., Tong, J., Zhang, F., Chu, F.: Intelligent fault identification in sample imbalance scenarios using robust low-rank matrix classifier with fuzzy weighting factor. Appl. Soft. Comput. 152, 111229 (2024)

    2024

  56. [65]

    IEEE Trans

    Yang, J., Zhang, D., Frangi, A.F., Yang, J.y.: Two-dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. Pattern Anal. Mach. Intell. 26(1), 131–137 (2004)

    2004

  57. [66]

    Yuan, Y., Chang, T.H., Sun, D.F., Toh, K.C.: A dimension reduction technique for large- scale structured sparse optimization problems with application to convex clustering. SIAM J. Optim. 32(3), 2294–2318 (2022)

    2022

  58. [67]

    arXiv preprint arXiv:2306.17369 (2023) Support matrix machine: exploring sample sparsity

    Yuan, Y., Lin, M., Sun, D.F., Toh, K.C.: Adaptive sieving: A dimension reduction technique for sparse optimization problems. arXiv preprint arXiv:2306.17369 (2023) Support matrix machine: exploring sample sparsity... 39

    2023 arXiv

  59. [68]

    Zhang, W., Liu, Y.: Proximal support matrix machine. J. appl. math. phys 10(7), 2268–2291 (2022)

    2022

  60. [69]

    Zhao, X.Y., Sun, D.F., Toh, K.C.: A Newton-CG augmented Lagrangian method for semidefinite programming. SIAM J. Optim. 20(4), 1737–1765 (2010)

    2010

  61. [70]

    IEEE Trans

    Zheng, Q., Zhu, F., Heng, P.A.: Robust support matrix machine for single trial EEG classification. IEEE Trans. Neural Syst. Rehabilitation Eng. 26(3), 551–562 (2018)

    2018

  62. [71]

    Pattern Recognit

    Zheng, Q., Zhu, F., Qin, J., Chen, B., Heng, P.A.: Sparse support matrix machine. Pattern Recognit. 76, 715–726 (2018)

    2018

  63. [72]

    Neurocomputing 275, 869–880 (2018)

    Zheng, Q., Zhu, F., Qin, J., Heng, P.A.: Multiclass support matrix machine for single trial EEG classification. Neurocomputing 275, 869–880 (2018)

    2018

  64. [73]

    Zhou, Z., So, A.M.C.: A unified approach to error bounds for structured convex opti- mization problems. Math. Program. 165, 689–728 (2017)

    2017

  65. [74]

    In: NIPS Workshop on Optimization in Machine Learning (OPT) (2015)

    Zimmert, J., de Witt, C.S., Kerg, G., Kloft, M.: Safe screening for support vector ma- chines. In: NIPS Workshop on Optimization in Machine Learning (OPT) (2015)

    2015

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.